Another new Yoyo application: Siever
Forum rules
- Dirk Broer
- Corsair
- Posts: 1962
- Joined: Thu Feb 20, 2014 11:24 pm
- Location: Leiden, South Holland, Netherlands
- Contact:
#1 Another new Yoyo application: Siever
A new application can be selected in the project preferences, Siever. 64-bit Windows only so far.
#2 Re: Another new Yoyo application: Siever
I gave this project a shot on my Linux PC and I'm disappointed by the amount of credit awarded for successful work. For example one task took over 8 hrs to crunch, 2,044.25 points were claimed and only 599.06 were given? Why?
-
- General Bitchin'
- Posts: 6371
- Joined: Wed Aug 15, 2018 1:15 pm
- Location: Huntly, Scotland
- Contact:
#3 Re: Another new Yoyo application: Siever
I was equally unimpressed with M-Queens https://tsbt.co.uk/forum/viewtopic.php?f=109&t=13798
I'll be off this project like snow off a dyke as soon as the FB sprint happens. I am only running it as I can bunker yoyo speculatively for a month ahead
I'll be off this project like snow off a dyke as soon as the FB sprint happens. I am only running it as I can bunker yoyo speculatively for a month ahead
I think this is fool-proof but could you just try it for me please? • There are 10 types of people in the world; those who understand binary, and those who don’t
#4 Re: Another new Yoyo application: Siever
I completely don't understand this whole BOINC credit system. Amicable Numbers pays shit loads per one WU, same with Collatz and probably PrimeGrid by looking at project's stats. Why not attract more users by paying bit more considering some tasks are very lengthy.davidBAM wrote: ↑Wed Jul 17, 2019 12:39 pm I was equally unimpressed with M-Queens https://tsbt.co.uk/forum/viewtopic.php?f=109&t=13798
I'll be off this project like snow off a dyke as soon as the FB sprint happens. I am only running it as I can bunker yoyo speculatively for a month ahead
Speaking of M-Queens, I didn't even try to finish single task. I was put off but time needed to complete WU and short deadlines.
I will only let Cruncher app to run on old Android phone. I don't get much points for it but at least "no maintenance" is required for this device.
-
- General Bitchin'
- Posts: 6371
- Joined: Wed Aug 15, 2018 1:15 pm
- Location: Huntly, Scotland
- Contact:
#5 Re: Another new Yoyo application: Siever
I totally agree Hal
I simply don't understand why all projects can't pay a consistent amount based on GFlopHrs - they already have a "Estimated computation size" in GFlops for both cpu and gpu projects. How hard can it be?
And, yes of course, a GPU will pay much higher than a CPU because it does more work (and costs a helluva lot more per-thread to buy and run)
I simply don't understand why all projects can't pay a consistent amount based on GFlopHrs - they already have a "Estimated computation size" in GFlops for both cpu and gpu projects. How hard can it be?
And, yes of course, a GPU will pay much higher than a CPU because it does more work (and costs a helluva lot more per-thread to buy and run)
I think this is fool-proof but could you just try it for me please? • There are 10 types of people in the world; those who understand binary, and those who don’t
#6 Re: Another new Yoyo application: Siever
[mention]Hal Bregg[/mention] this is an ancient topic and has been going around every since GPU crunching 1st came out. Credit Screw, which Yoyo uses, was David Anderson's approach to "socialized" crunching. If he totally had his way it wouldn't matter if you were running a single Pentium computer or a 2990WX Threadripper you would get paid the same
There are projects I only run if I have to because of the credits (ODLK, ODLK1) immediately come to mind.
There are projects I only run if I have to because of the credits (ODLK, ODLK1) immediately come to mind.
#7 Re: Another new Yoyo application: Siever
Bear in mind Bryan that I am fairly new to world of BOINC and to be honest I didn't do any deeper research into credit system. I will sound stupid or unthought but here it isBryan wrote: ↑Wed Jul 17, 2019 2:22 pm @Hal Bregg this is an ancient topic and has been going around every since GPU crunching 1st came out. Credit Screw, which Yoyo uses, was David Anderson's approach to "socialized" crunching. If he totally had his way it wouldn't matter if you were running a single Pentium computer or a 2990WX Threadripper you would get paid the same
There are projects I only run if I have to because of the credits (ODLK, ODLK1) immediately come to mind.
You said that there would be no much difference between Pentium and Threadripper which would suggest that there is no difference in computational power between those two. So if there is no difference, it would indicate that benchmarking employed in BOINC is outdated or not written properly.
- scole of TSBT
- Boinc Major General
- Posts: 5980
- Joined: Mon Feb 03, 2014 2:38 pm
- Location: Goldsboro, (Eastern) North Carolina, USA
#8 Re: Another new Yoyo application: Siever
Hal, there are no stupid questions. Project admins have the option to use the Credit New method or using other options. Some projects award credit generously while others not so much. Credit New is an odd animal. On some projects it will pay handsomely for a short period of time but after that it will level off to "normal" credit. As for the Pentium Vs. Threadripper, Bryan meant if the David Anderson, one of the founders of BOINC, had his way there would be little credit difference but there was a little sarcasm/exaggeration aimed toward David Anderson.Hal Bregg wrote: ↑Wed Jul 17, 2019 4:57 pmBear in mind Bryan that I am fairly new to world of BOINC and to be honest I didn't do any deeper research into credit system. I will sound stupid or unthought but here it isBryan wrote: ↑Wed Jul 17, 2019 2:22 pm @Hal Bregg this is an ancient topic and has been going around every since GPU crunching 1st came out. Credit Screw, which Yoyo uses, was David Anderson's approach to "socialized" crunching. If he totally had his way it wouldn't matter if you were running a single Pentium computer or a 2990WX Threadripper you would get paid the same
There are projects I only run if I have to because of the credits (ODLK, ODLK1) immediately come to mind.
You said that there would be no much difference between Pentium and Threadripper which would suggest that there is no difference in computational power between those two. So if there is no difference, it would indicate that benchmarking employed in BOINC is outdated or not written properly.
#9 Re: Another new Yoyo application: Siever
Well maybe just a little bit of sarcasm
Hal, the original intent of the credit system was that it would be the same cross-project. Meaning regardless of which project you ran computer X would make about the same credits on any of them. They had a "reference" machine and the credits on the projects all used that type of machine to establish the credit level so things remained relatively on par.
Then along came GPUs and things deteriorated fairly quickly. The GPUs could out produce CPUs by a big margin so the credits granted were raised accordingly. The people who had GPUs thought it was great and the people who didn't have GPUs screamed bloody murder.
The ultimate was a project called Bitcoin Utopia that used dedicated mining asics and paid credits accordingly. It was no problem to get 1B credits/day That single project basically destroyed any thoughts of parity across all projects.
It costs the projects nothing to grant credits ... either high or low. Since I'm a badge hunter and try to reach certain personal goals (based on credits) then I try to avoid the really low paying projects. But by the same token when I reach a goal I stop crunching high paying projects also.
Your question was NOT dumb by any means. There just isn't a SMART answer to it ... it has been an on going hot topic since the beginning. There will never be cross project parity. It is what it is.
Hal, the original intent of the credit system was that it would be the same cross-project. Meaning regardless of which project you ran computer X would make about the same credits on any of them. They had a "reference" machine and the credits on the projects all used that type of machine to establish the credit level so things remained relatively on par.
Then along came GPUs and things deteriorated fairly quickly. The GPUs could out produce CPUs by a big margin so the credits granted were raised accordingly. The people who had GPUs thought it was great and the people who didn't have GPUs screamed bloody murder.
The ultimate was a project called Bitcoin Utopia that used dedicated mining asics and paid credits accordingly. It was no problem to get 1B credits/day That single project basically destroyed any thoughts of parity across all projects.
It costs the projects nothing to grant credits ... either high or low. Since I'm a badge hunter and try to reach certain personal goals (based on credits) then I try to avoid the really low paying projects. But by the same token when I reach a goal I stop crunching high paying projects also.
Your question was NOT dumb by any means. There just isn't a SMART answer to it ... it has been an on going hot topic since the beginning. There will never be cross project parity. It is what it is.
#10 Re: Another new Yoyo application: Siever
Sorry guys for going off topic. And thanks for sharing your thoughts.
Like Bryan said above "There just isn't a SMART answer to it" so let's close this topic and get back to crunching.
Like Bryan said above "There just isn't a SMART answer to it" so let's close this topic and get back to crunching.